数据加载中.....
|
jsp.display-item.identifier=請使用永久網址來引用或連結此文件:
https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/35911
|
| 题名: | Development of watermarking scheme for halftone images |
| 其它题名: | 半色調影像浮水印系統之研究 |
| 作者: | 吳業寬;Wu, Yeh-kuang |
| 贡献者: | 淡江大學電機工程學系博士班
謝景棠;Hsieh, Ching-tang |
| 关键词: | 浮水印;半色調;粒子群最佳化;支援向量機;人眼視覺系統;Watermark;Halftone Image;particle swarm optimization;Support Vector Machine;Human visual system |
| 日期: | 2009 |
| 上传时间: | 2010-01-11 07:19:04 (UTC+8) |
| 摘要: | 數位多媒體資訊被廣泛地運用於我們的日常生活當中，人們可以輕易地在未經擁有者同意的情況下，恣意地複製與傳播。數位浮水印技術正是一個解決此問題的好方法，實行數位多媒體版權的認證與智慧財產權的保護。數位浮水印技術中，根據嵌入浮水印的目的不同，可分為強健型與易碎型浮水印系統，每種系統都有其一定的限制與考量；本論文中，我們將分別針對此兩系統提出新的策略，應用於數位影像中，同時並將其擴展於數位半色調影像中。相較於灰階影像，數位半色調影像僅需使用極少的記憶體儲存空間，不僅利於網路上的傳輸，亦能產生近似灰階的效果，遂其廣泛地利用在網路與列印之相關應用中。
本論文中，我們將先簡介強健型與易碎型浮水印系統之特性。強健型浮水印可運用於智慧財產權之保護，而易碎型浮水印可運用於資料之保護與認證。
本論文的第二個主軸將探討應用於半色調影像中之浮水印系統。首先，我們提出了一個多目標之半色調影像浮水印系統，多目標即代表此浮水印系統同時擁有強健型浮水印及易碎型浮水印之特性。藏有浮水印之半色調影像必須能同時具有透視性與強健性，在本系統中，我們引用粒子群最佳化演算法，藉以找到最佳化之點陣式矩陣，並經由數個矩陣之選擇，嵌入強健型浮水印系統。同時為了不影響強健型浮水印之強健性，我們將易碎型浮水印嵌入於強健型浮水印之鎖鑰中。
經由模擬，我們發現上述之方法，並不能抵抗影像尺寸縮小放大之攻擊；有鑒於此，我們提出了第二種半色調影像浮水印系統，稱之為混合型浮水印系統，此系統結合了上述之多目標系統，並於嵌入了另一個浮水印於頻率域上，藉以增加浮水印系統之強健度。同時為了衡量半色調影像浮水印之品質，我們利用粒子群最佳化演算法，提出了人眼視覺濾波器，應用於PSNR-影像衡量法中，經由人眼視覺濾波器可以將半色調影像較為準確地近似至灰階影像。
然而，上述之方法，在其浮水印抽取步驟中，皆須使用此多個半色調矩陣。最後，我們試著提出一進階型浮水印系統，此系統將簡化浮水印抽取時所需要的事前資訊，以確保浮水印系統之隱密性，並避免竄改者恣意產生未經許可，確具有浮水印之半色調影像。同時，對於列印與掃描攻擊，我們提出一套自動化同步系統應用於浮水印抽取系統中，包含所有相關之前處理作業。
As audio, video, images and other works become available in digital form, the ease with which perfect copies can be made, may lead to large-scale unauthorized copying which might undermine the music, picture, book, and software publishing industries. Watermarking has been considered for many copy prevention and copyright protection applications. Watermarking is the practice of hiding a message about an image, audio chip or meaningful logo within that work itself for protecting copyright. Two types of watermarking can be identified depends on specific application-driven requirements: (a) robust watermarking, and (b) fragile watermarking. For the aforementioned schemes, each watermarking scheme has its own considerations and limitations. In this dissertation, we are trying to develop robust and fragile watermarking schemes respectively and explore the combined watermarking schemes for the halftone images. Halftone images are widely used in the printing of books, magazines, newspapers and computer printers. Although there are only 2 tones, halftone images look like the original multi-tone images when viewed from a distance.
The first issue of this dissertation is to introduce the robustness and fragility of the watermarking system for digital gray-level image. The robust watermarking scheme is used for copyright protection and fragile watermarking scheme is used for content authentication. The second issue of this dissertation focuses on embedding the watermarks in the halftone images. The optimized watermarking scheme for halftone images based on dithering method is proposed. The multi-purpose watermarking is proposed for not only protecting copyright but also authenticating content. The optimized dither cells for watermarks embedding would be generated by the particle swarm optimization (PSO) with high transparency. However, in the experiments, we find that the proposed multi-purpose watermarking is not robust to the scaling attacks.
In the third issue of the dissertation, we would try to find the more robust watermarking scheme and discuss the image quality criterion for halftone images. In the method, the frequency domain based watermarks would be embedded to the watermarked images in order to enhance the robustness for scaling attack and call it hybrid watermarking scheme. And the human visual filter for the image quality criterion in halftone images is generated by PSO.
The last issue of this dissertation is to explore the advanced method for the proposed robust watermarking system for halftone images. The section aims at enhancing the privacy of the watermarking scheme that only one dither cell is necessary in the watermark extracting system. The support vector machine is adopted as the classification function in the watermark extracting stage, and the automatic resynchronization procedure for the watermark extraction stage with print-and-scan attacks is proposed to resist scan-and-print attacks. |
| 显示于类别: | [電機工程學系暨研究所] 學位論文
|
文件中的档案:
| 档案 |
大小 | 格式 | 浏览次数 |
|---|
| 0Kb | Unknown | 372 | 检视/开启 |
|
在機構典藏中所有的数据项都受到原著作权保护.
|
